Automated detection of sleep apnea via predictive model based on patient health and medical data
Principal Investigator: Dr Niranjan Sridhar
Approved Research ID: 48716
Approval date: June 19th 2020
Our goal is to be able to extract predictive factors to construct a sleep apnea diagnosis predictive model for identifying people that are at high risk of sleep apnea and co-morbidities associated with sleep apnea. Research has shown that majority of the US population alone has about 80% undiagnosed OSA individuals. Our screening diagnosis algorithm using this predictive model based on health and medical data hopes to reach undiagnosed OSA patients at scale to reduce patient burden of clinical path of diagnosis for those that are at high risk of OSA. Identified high risk OSA individuals will be offered the opportunity to be educated about OSA to long-term treatment adherence, in order to improve patient sleep quality in relation to clinical outcomes, especially in co-morbid and high risk and high-cost populations. For examples, cardiovascular and metabolic diseases.